Re: PowerPC bootdisks for potato release.

2000-02-16 Thread Brad Midgley
an important caveat: for all 3 booters in my cd image (bootx, yaboot, miboot) i specify root=1 in the kernel arguments. you'll probably want to remove that. (specifying root allows /linuxrc to be executed; the value '1' is used by linuxrc to determine that real-root-dev is backward endian, and lin

Re: PowerPC bootdisks for potato release.

2000-02-16 Thread Brad Midgley
> I am beginning to think that the better option is a boot CD and a > second ISO format CD which is not bootable, once your booted you > remove the boot CD and use the binary CDs which are in ISO format. I > > sounds like a plan to me, comments? if you'd like, you can use the image i've been wor

Re: PowerPC bootdisks for potato release.

2000-02-16 Thread Ethan Benson
On Tue, Feb 15, 2000 at 04:24:22PM -0500, [EMAIL PROTECTED] wrote: > The other possibility is the 'hybrid' ISO/HFS format -- will Macs boot that > or is it a no-go? > > The nice thing about this is that you'd be able to see the files on the CD > from MacOS and Linux equally without having to have

Re: PowerPC bootdisks for potato release.

2000-02-16 Thread Ethan Benson
On Tue, Feb 15, 2000 at 08:23:56PM -0800, Brad Midgley wrote: > > The other possibility is the 'hybrid' ISO/HFS format -- will Macs boot that > > or is it a no-go? > > my experience is that new world macs' OF will see the iso side and fail to > boot because the iso driver is so poor. i haven't tri

Re: PowerPC bootdisks for potato release.

2000-02-16 Thread Brad Midgley
> The other possibility is the 'hybrid' ISO/HFS format -- will Macs boot that > or is it a no-go? my experience is that new world macs' OF will see the iso side and fail to boot because the iso driver is so poor. i haven't tried this recently with yaboot though. if there's a way to tell OF to try

Re: PowerPC bootdisks for potato release.

2000-02-15 Thread Geert Uytterhoeven
On Tue, 15 Feb 2000 [EMAIL PROTECTED] wrote: > On Fri, Feb 11, 2000 at 10:56:24AM +0100, Benjamin Herrenschmidt wrote: > > For the "linux" part of the CD, you have several solutions: You can > > simply have an ext2 image in a big file on the HFS partition and mount it > > via loopback (slow). You c

Re: PowerPC bootdisks for potato release.

2000-02-15 Thread james
On Fri, Feb 11, 2000 at 10:56:24AM +0100, Benjamin Herrenschmidt wrote: > For the "linux" part of the CD, you have several solutions: You can > simply have an ext2 image in a big file on the HFS partition and mount it > via loopback (slow). You can also probably add ext2 partitions using the > appl

Re: PowerPC bootdisks for potato release.

2000-02-14 Thread Brad Midgley
don't be tempted to put in a rock-ridge partition or make it hybrid though. i found that new-world machines' OF ignores the hfs data if it can find an iso partition yet they can't properly boot from the iso partition. ben wrote: > For the "linux" part of the CD, you have several solutions: You can

Re: PowerPC bootdisks for potato release.

2000-02-14 Thread Brad Midgley
i used toast to create a bootable image, but didn't let it optimize so there's a lot of free space in the hfs partition. now i can loop-mount it under linux (hfs finds the right partition in the image to mount) to fix up the content and i don't have to use toast again (this part does need to be ve

Re: PowerPC bootdisks for potato release.

2000-02-11 Thread Ethan Benson
On Fri, Feb 11, 2000 at 10:56:24AM +0100, Benjamin Herrenschmidt wrote: > > If we _really_ want to make something absolutely and completely stripped > of any Apple code, miBoot should be turned into a fake CD driver burned > in the disk's partition map. This is possible but more complex and I > di

Re: PowerPC bootdisks for potato release.

2000-02-11 Thread Geert Uytterhoeven
On Thu, 10 Feb 2000, Ethan Benson wrote: > yaboot is an OpenFirmware executable which loads the linux kernel from > the ext2 filesystem. Aha, any chance it also works on CHRP boxes. Hartmut, did you ever try? Gr{oetje,eeting}s, -- Geert Uytterhoeven -- Linux/{m68k~Amiga,PPC~CHRP} -- [EMAIL PROT

Re: PowerPC bootdisks for potato release.

2000-02-11 Thread Benjamin Herrenschmidt
On Thu, Feb 10, 2000, Ethan Benson <[EMAIL PROTECTED]> wrote: >doesn't Miboot also require non-free Apple bootblock code in the 1K >bootblock of the HFS partition? Hum... the status of this bootblock code is difficult to determine. It's only a few bytes of 68k assembly that calls the ROM _InitFS

Re: PowerPC bootdisks for potato release.

2000-02-11 Thread Ethan Benson
On Thu, Feb 10, 2000 at 11:41:12AM +0100, Hartmut Koptein wrote: > > Is Yaboot required ? I assume this is the equivalent of lilo on a x86 > > machine. > > yaboot is more like loadlin; quik is the equivalent for lilo. actually no, BootX is more like loadlin, loadlin: a DOS application which

Re: PowerPC bootdisks for potato release.

2000-02-11 Thread Ethan Benson
On Thu, Feb 10, 2000 at 01:18:06AM -0500, Daniel Jacobowitz wrote: > On Thu, Feb 10, 2000 at 10:34:14AM +1100, Brendan J Simon wrote: > > > > Are the boot disks going to be "ready" for the upcoming relesae of > > powerpc-potato in February (it is still February isn't it ?) ? > > Looks like it. >

Re: PowerPC bootdisks for potato release.

2000-02-11 Thread Ethan Benson
On Thu, Feb 10, 2000 at 10:34:14AM +1100, Brendan J Simon wrote: > > Are the boot disks going to be "ready" for the upcoming relesae of > powerpc-potato in February (it is still February isn't it ?) ? > > I want to buy a PowerBook but am waiting for Debian to be officially > released and with boo

Re: PowerPC bootdisks for potato release.

2000-02-10 Thread Hartmut Koptein
> Is Yaboot required ? I assume this is the equivalent of lilo on a x86 > machine. yaboot is more like loadlin; quik is the equivalent for lilo.

Re: PowerPC bootdisks for potato release.

2000-02-10 Thread Daniel Jacobowitz
On Thu, Feb 10, 2000 at 10:34:14AM +1100, Brendan J Simon wrote: > > Are the boot disks going to be "ready" for the upcoming relesae of > powerpc-potato in February (it is still February isn't it ?) ? Looks like it. > Is Yaboot required ? I assume this is the equivalent of lilo on a x86 > machi